We propose an innovative, integrated, multi-component, multi-level, multi-setting (MMM) approach to treating overweight and obese children (BMI > 85th percentile). The experimental MMM intervention is based on successful past research and ongoing community based participatory research (CBPR), and is designed to overcome barriers to existing treatment models. It includes: a theory-based community center team sports program designed specifically for overweight & obese children; a home-based family intervention to reduce screen time and to alter the home food/eating environment; and a primary care provider behavioral counseling intervention linked to the community and home interventions. After two years of formative research and pilot studies, we will evaluate our intervention in a randomized controlled trial. 240 families with overweight and obese children will be recruited through primary care pediatric clinics serving low-income, ethnically-diverse patient populations in East Palo Alto and Redwood City, California. Eligible participants will be referred by their primary medical care providers to after school programs conducted by their local YMCAs and Boys and Girls Clubs. After completing baseline assessments and enrolling in the study, families will be randomized to standard clinical care and the standard community center programs, or to the experimental MMM intervention for 36 months. All participants will complete assessments at baseline, 12 months, 24 months and 36 months. The primary outcome measure is change in BMI over the entire course of the study. Secondary outcomes include changes to 12 and 24 months, and changes in waist circumference, triceps skinfold thickness, blood pressures, heart rate, fasting lipoprotein levels, glucose and Insulin, hsCRP, physical activity (accelerometery), screen time and other sedentary behaviors, dietary intake, weight concerns, and depressive symptoms. DNA will be collected to identify genetic moderators of intervention responsiveness. We will compare the MMM intervention and standard care control groups using random regression models to take full advantage of all longitudinal data. Primary Hypothesis: Compared to standard care controls, children randomized to our multi-component, multi-level, multi-setting (MMM) intervention will significantly reduce their body mass index. We will also examine potential baseline biological, psychological, social and environmental moderating and mediating variables, cost-effectiveness and public policy implications.